Effective labelling is difficult, but safety really does matter

The recent letter discussing the Poggendorf effect (Nott. Anaesthesia 2001; 56: 917) is helpful in understanding some of the complex psychological aspects of labelling and safety. His suggestion that a second person must also read the label before the administration of any drug is certainly of value, but raises a number of issues. The first concerns resource – a survey of New Zealand anaesthetists found that in only 1.5% of circumstances was a second person routinely available throughout the anaesthetic to check drugs before their administration [1]. The second, more important issue, however, relates to human fallibility. We have received a number of routine audit reports documenting errors made by nursing staff who have used a two-person check, but still got it wrong. Part of the problem seems to be the suggestibility of human beings. If the second person is shown a label and asked if it is indeed drug X, then there is a chance that he or she will read the label as drug X, even if it is in fact a similarly named but different drug. We have developed a new system designed to reduce the propensity for error during drug administration in anaesthesia [2]. One feature of this system is the use of large legible labels that include bar codes. When a syringe with such a label is scanned by the barcode reader just before administration, a computer speaks the name of the drug as a final check of the syringe's contents. The computer is of course not open to suggestion and our system has already demonstrated its ability to facilitate anaesthetists' efforts to avoid drug error. In this way, we have embodied the principle of the two-person check, in a manner that utilises a less expensive resource than a trained health professional, and that capitalises on the strength of machines in an area in which humans are prone to error. The approach also makes possible the inclusion of various alarms for expired drugs and allergy alerts. No safety system can be expected to eliminate errors and accidents entirely [2, 3]. We therefore agree with Dr␣Nott's statement that colour coding by class of drug will not eliminate error on its own. Neither will the use of the bar-coded ‘two-person check’ described above. However, we believe that each would be a significant improvement on the status quo, and both should be incorporated into a wide-ranging approach to error reduction, which includes (so far as possible) all aspects of the drug administration environment in theatre and indeed beyond [2, 4, 5]. The use of colour-coded (and bar-coded) user-applied syringe labels does nothing to address the problem of ampoule labelling. We have previously suggested that manufacturers should supply injectable drugs in appropriately labelled prefilled syringes to remove one error-prone step from the process of drug administration [4]. In reality, this will be a long time coming, and the current state of labelling on most ampoules can only be described as appalling. Figure 14 shows two ampoules from a drug trolley at our hospital, which were recently involved in a substitution near miss. These ampoules are compliant with the current NHS specification for ampoule labelling [6]. Yet, despite the fact that the figure shows the drug brand names facing forward (in the largest text on the ampoule), the ampoules remain very difficult to read and distinguish from one another. The black writing on clear glass is hard to read because the writing on the other side of the ampoule shows through and obscures critical information. The NHS specification suggests the use of a yellow background for the lettering on ampoules to increase legibility – but obviously it is simpler and allowable for manufacturers not to bother. The bands at the base of these ampoules may look like an identification aid such as the rings sometimes used to indicate the length of action of a drug [7], but they are in fact specialised barcodes (without expiry dates) used by the manufacturer (personal communication, Abbott Laboratories). We use custom-made prefilled syringes where practical in our new system [2, 4]. In addition we have designed a ‘flag’ or supplementary label attached to the ampoule by a licensed pharmaceutical manufacturer with a loop of clear plastic [2]. The flag label adds the drug name, drug class, its class-specific colour cue and a barcode, all in a highly legible manner, without obscuring any of the ampoule's original details (Fig. 15). When the contents of the ampoule are drawn up, the self-adhesive flag label can be removed and attached to the syringe as part of the drawing-up process. Consistent colour standards for user-applied syringe labels in anaesthesia already exist in Australia, New Zealand, Canada and the United States (discussed in [2]). In addition, ampoule standards also exist in Canada and the US, which allow the use of a colour-coded border, consistent with the aforementioned syringe label standards, around the␣‘critical information panel’– thus permitting the use of consistent ampoule and syringe label colour cues [8, 9]. Ampoule colour coding by class of drug was suggested 19 years ago, but remains undeveloped [7]. While it may be␣convenient from an administrative, bureaucratic or manufacturing perspective to have different standards for ampoules and syringes, this makes no sense in terms of the ergonomics and safety of a clinician administering a drug to a patient. There is no doubt that getting the detail of effective labelling standards right is difficult, but if safety really does matter [10], then these difficulties must be overcome. Resolution of these problems is long overdue.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.